PROJECT SUMMARY To understand neuropsychiatric disorders behavioral phenotypes, have to be linked to altered brain function and brain circuits. However, most neuropsychiatric disorders show a range of phenotypes and high variability making the association with neurological correlates difficult. Development of animal models of neuropsychiatric disorders such as autism, have been facilitated by gene discovery through linkage analysis. Pharmacological treatments also model aspects of these disorders in rodents. However, comprehensive behavioral and physiological analysis is very time consuming. Moreover, common behavioral and physiological testing is performed in specialized apparatuses requiring handling of the rodents which in turn can affect their performance and lead to experimenter and environment induced variability. This it is difficult to observe multiple measures such as behavioral and neural activity in the same animal without perturbation and can capture developmental trajectories of disease progression and/or therapeutic intervention. We here aim to solve this problem by developing a new, flexible platform in which mice can be continually assessed within their home cage environments with a barrage of behavioral tests as well as neurophysiological measures (brain functional imaging) and optogenetic intervention. Our system will include autonomous monitoring and control of brain activity within these home cages in the context of stimuli used to assess auditory sensory processing. The combination of the home cage environments with auditory learning paradigms provides a single platform for both training, assessment, and potential therapeutic manipulation. Our strategy will employ a collaborative effort to build home cages for up to 10 mice/cage, which can be housed continually for up to a 1-year period. Over this time, the animals will initiate self-directed trials of brain imaging, as well as auditory processing assessments. A core feature of our system is the usage of relatively small footprint open source, Linux-based computer tools employing single board Raspberry Pi and Arduino computers. These systems will work together to enable the flexible home cage training and assessment system. The first goals are to refine the hardware and software to enable cross-laboratory collaboration and then wider dissemination of these tools to the broader neuroscience community. We will test the home cages on specific lines of mice which have previously been shown to model autistic-like behaviors and uncover co-variation of behavioral, cognitive, and physiological deficits within and across the various models.